Search has evolved.
We started with simple keyword search, improved it with full-text search, and are now entering the era of semantic search powered by vectors.
But here’s the reality: most software systems today are still stuck somewhere between keyword and full-text.
In this post, we’ll walk through:
- Keyword vs Full-Text vs Vector search
- Why you need vector databases — even if you already use PostgreSQL full-text search
- Similarity algorithms: cosine, dot product, Euclidean — with visuals
- A full .NET 9 RAG implementation using PostgreSQL +
pgvector+ Ollama
